1. Field of the Invention
The present invention relates to an image data compression apparatus which compresses image data for each predetermined block, and an image processing apparatus which performs image processing and restoration processing for the compressed data. The present invention more particularly relates to an image processing apparatus which performs composition processing for the compressed data.
2. Description of the Related Art
Conventionally, a JPEG scheme which uses discrete cosine transformation and a scheme which uses wavelet transformation are widely employed as color still image compression schemes. Encoding schemes of these kinds generally encode image data for each predetermined block (for example, for each unit of 8×8 or 16×16 pixels), and perform discrete cosine transformation, quantization, and entropy coding, thereby realizing high compression efficiency. Since encoding schemes of these kinds use variable-length coding, the code length changes for each image data to be encoded. To perform discrete cosine transformation processing for each block at high speed, not only a high-speed arithmetic unit and dedicated hardware but also a buffer memory is necessary.
Also, the number of pixels which require image processing typified by composition processing has dramatically increased together with an increase in resolution, which also increases the processing load. To execute image processing of referring to pixel data and converting its value for compressed data having undergone the above-mentioned image compression, decoding the compressed data becomes necessary. In other words, image processing typified by composition processing cannot be performed for compressed data intact, so the decoding of compressed data inevitably becomes necessary. This makes it necessary to process all pixels in high-resolution data for each pixel, thus prolonging the processing time.
Japanese Patent Laid-Open No. 2000-358194 proposes an image composition apparatus for preventing prolongation of the image composition processing time with an increase in resolution. According to Japanese Patent Laid-Open No. 2000-358194, reduced images of a plurality of image data to be composed are generated, and extraction of overlapping regions and alignment are performed. Based on the information obtained by these operations, the image data are composed.
However, in the method disclosed in Japanese Patent Laid-Open No. 2000-358194, the processing time prolongs generating reduced images since reduced images of image data that are to be composed must be generated. Also, since composition processing is performed for image data before reduction, the memory capacity for temporary storage increases.